As many countries struggle with large deficits, some are looking to respond by cracking down on tax evasion. For example, Greece’s medium-term fiscal program assumes an improvement in tax collection of 1.5% of GDP. Achieving such a result generally involves increasing deterrence, by upping enforcement and raising penalties.

To increase the chances of detecting evasion, the Greek government created a powerful, autonomous office modelled on the US Internal Revenue Service. The US has recently extended its system of information reporting; as of 2011 it requires payment processors (such as credit card companies) to inform the IRS of the receipts of businesses that use their payment systems, enabling the IRS to match these reports to reported business receipts.

Tax evasion matters

Much recent research has explored the efficacy of alternative enforcement strategies. Pomeranz (2012) analyses the role of information in tax enforcement in the case of the Chilean value added tax (VAT) through two randomized field experiments involving over 400,000 Chilean firms. This is an especially apt setting because claims that the VAT facilitates tax enforcement – by generating a paper trail on transactions between firms – have contributed to its widespread adoption. It is now used in over 160 countries.

Pomeranz finds that the paper trail acts as a substitute to a firm's own audit risk. A message announcing increased tax enforcement has a much smaller effect on the reporting of transactions that are already covered by a paper trail. A second experiment shows that the paper trail leads to spillovers that create important multiplier effects in tax enforcement. The impact of a random audit announcement is transmitted up the VAT chain, increasing compliance by firms' suppliers.

These findings confirm that when evasion is taken into account, significant differences emerge between taxes that are equivalent in standard models but generate different information on taxable transactions. A retail sales tax and a VAT are both consumption taxes and are equivalent in standard analytical models. However, the details of the collection and remittance systems differ. The retail sales tax collects all revenue from retail businesses, for which there is no feasible corroborating party (since consumers are too numerous), while the VAT is collected from all businesses along the production and distribution chain (Baldwin 2007).

Recent research

We take up this theme in a recent paper entitled “Do the Laws of Tax Incidence Hold? Point of Collection and the Pass-Through of State Diesel Taxes.” We examine state taxes on diesel fuel and empirically demonstrate that the identity of the tax remitter—the party along the production and distribution chain responsible for remitting the tax to the government — affects both amount of taxes collected and the incidence of taxes. In sharp contrast to the textbook case, the efficacy and distributional consequence of tax collection depend on who remits the taxes. As with the VAT, the tax authority’s choice of where to assign responsibility is critical.

Our analysis hinges on the fact that states differ in the stage of the supply chain responsible for remitting the tax (none require final consumer remittance). Over time states have moved the point of tax collection higher in the supply chain. In the early 1980s, states were almost evenly split between collecting taxes from retail stations and collecting from intermediate wholesale distributors. Over the subsequent 20 years, many states have moved away from collecting from retail stations and towards collecting from prime suppliers.


We find that the rate of pass-through of diesel taxes to retail prices is dependent on the identity of the remitting party. Retail diesel prices are higher, and diesel taxes exhibit greater pass-through to retail prices, in states where the point of collection is upstream – at the distributor or prime-supplier level – rather than at the level of the retail station. Furthermore, we find evidence suggesting that, ceteris paribus, states collect less tax revenue when taxes are collected at the retail level. These three pieces of evidence suggest that in the case of diesel taxes, collecting taxes from upstream parties reduces evasion and suggests that retailers have more ability to evade taxes than firms at higher levels of the supply chain.

Lending credence to this result, we trace the impact of collecting at the prime-supplier level through the supply chain. An increase in the tax raises the wholesale price in supplier-remitting states, but not in retailer-remitting states. Moreover, conditional on the wholesale price, supplier remittance has no effect on pass-through. In other words, the effect of a tax increase on retail prices in supplier-remitting states can be entirely explained by their effect on wholesale prices.

Concluding remarks

Our findings might not be surprising to some in the tax-administration community. However, they contradict the standard economic theory of taxation. The standard theory asserts that the identity of the party that remits a tax liability has no effect on how much revenue is raised, nor does it determine which party bears the burden of the tax. These results hold in under very general theoretical conditions.

But the ‘general conditions’ assume the absence of tax evasion, and thus apply to a world vastly different than Greece or the US – or any other country in the world.


Baldwin, Richard (2007) "VAT fraud part 1",, 14 June.

Kopczuk, Wojciech, Justin Marion, Erich Muehlegger and Joel Slemrod (2013), “Do the Laws of Tax Incidence Hold? Point of Collection and the Pass-Through of State Diesel Taxes,” NBER Working Paper No. 19410, September, 2013.

Pomeranz, Dina, "No Taxation without Information: Deterrence and Self-Enforcement in the Value Added Tax," Harvard Business School Working Paper, No. 13–057, December 2012.


2,205 Reads